remote sensing

Letter

Volcanic Cloud Top Height Estimation Using thePlume Elevation Model Procedure Applied toOrthorectified Landsat 8 Data. Test Case: 26 October2013 Mt. Etna Eruption

Marcello de Michele 1,*, Daniel Raucoules 1 , Stefano Corradini 2, Luca Merucci 2 ,Giuseppe Salerno 3 , Pasquale Sellitto 4 and Elisa Carboni 5

1 Bureau de Recherches Géologiques et Minières, 3 av. C. Guillemin, 45000 Orléans, France;d.raucoules@brgm.fr

2 Istituto Nazionale di Geofisica e Vulcanologia, Sez. Roma Via di Vigna Murata, 00100 Rome, Italy;stefano.corradini@ingv.it (S.C.); luca.merucci@ingv.it (L.M.)

3 Istituto Nazionale di Geofisica e Vulcanologia, Sez Catania, Piazza Roma, 95100 Catania, Italy;giuseppe.salerno@ingv.it

4 Laboratoire Interuniversitaire des Systèmes Atmosphériques, Université Paris-Est, 94000 Créteil, France;pasquale.sellitto@u-pec.fr

5 COMET, Atmospheric, Oceanic and Planetary Physics, University of Oxford, Clarendon Laboratory,Parks Road, Oxford OX1 3PU, UK; elisa.carboni@physics.ox.ac.uk

* Correspondence: m.demichele@brgm.fr

Received: 22 January 2019; Accepted: 28 March 2019; Published: 2 April 2019�����������������

Abstract: In this study, we present a method for extracting the volcanic cloud top height (VCTH) as aplume elevation model (PEM) from orthorectified Landsat 8 data (Level 1). A similar methodologywas previously applied to raw Landsat-8 data (Level 0). But level 0 data are not the standardproduct provided by the National Aeronautics and Space Administration (NASA)/United StatesGeological Survey (USGS). Level 0 data are available only on demand and consist on 14 data stripesmultiplied by the number of multispectral bands. The standard product for Landsat 8 is the orthoimage, available free of charge for end-users. Therefore, there is the need to adapt our previousmethodology to Level 1 Landsat data. The advantages of using the standard Landsat productsinstead of raw data mainly include the fast -ready to use- availability of the data and free access toregistered users, which is of major importance during volcanic crises. In this study, we adapt thePEM methodology to the standard Landsat-8 products, with the aim of simplifying the procedurefor routine monitoring, offering an opportunity to produce PEM maps. In this study, we presentthe method. Our approach is applied to the 26 October 2013 Mt. Etna episodes comparing resultsindependent VCTH measures from the spinning enhanced visible and infrared imager (SEVIRI) andthe moderate resolution imaging spectroradiometer (MODIS).

Keywords: volcanic cloud; Landsat 8; elevation model

1. Introduction

In volcanology, the volcanic cloud-top height (VCTH) is one of the most critical parameters toretrieve. It affects the quantitative estimation of volcanic cloud ash and gases parameters [1–3], the masseruption rate needed for the transport and deposition models [4–6] and the definition of the mostdangerous zone for air traffic. Exploiting their global coverage (in time and space), satellite sensorsoffer the unique possibility for an effective monitoring of VCTH. In recent years, many techniqueshave been developed exploiting the dark pixel brightness temperature [2], the CO2 [7,8] and O2 [9,10]

Remote Sens. 2019, 11, 785; doi:10.3390/rs11070785 www.mdpi.com/journal/remotesensing

Remote Sens. 2019, 11, 785 2 of 12

absorption bands, the radio occultation [11] and backward trajectory modelling [12]. Among thedifferent techniques (for a complete review of advantages and drawbacks please refer to [13]) andsatellite active systems as CALIPSO [14], several algorithms have been developed exploiting theparallax between remote sensing measurements collected by different views of the same object byusing one or more instruments. The use of the parallax was introduced by Prata and Turner [15] usingthe dual view of the along track scanning radiometer (ATSR) and developed further by Mims et al. [16],Nelson et al. [17], and Flower and Kahn [18] using dedicated multiangle imager spectro radiometer(MISR) measurements. Zakšek et al. [19], Corradini et al. [20], and Merucci et al., [13] developedalgorithms based on the combined use of polar-geostationary, ground based-geostationary andgeostationary-geostationary measurements respectively.

De Michele et al. [21] generalized the method based on small parallax for virtually all pushbroom sensor data. The extraction of the VCTH in the form of a plume elevation model from thehigh-resolution push broom operational land imager (OLI) sensor on board a Landsat-8 satellite,has been demonstrated starting from raw data. The main idea expressed in [21] is that the physicaldistance between the panchromatic sensor (PAN) and the multi-spectral sensors (MS), both onLandsat-like satellites, yields a baseline and a time lag between the PAN and MS image acquisitionsduring a single passage of the satellite. This information can be used to extract a spatially detailed mapof VCTH from virtually any multi spectral push broom system, called a plume elevation model (PEM).

The main difficulty of the data processing comes from the fact that one Landsat image is composedof 14 focal plane modules (FPMs) arranged in the so called ’staggered’ geometry, which makes the jointretrieval of plume velocities and heights challenging. De Michele et al. [21] addressed this problemby reconstructing a new OLI image starting from the raw OLI data stripes (courtesy of NationalAeronautics and Space Administration (NASA), pers. comm.). However, the raw Landsat-8 data arenot the standard Landsat-8 products provided by NASA/ United States Geological Survey (USGS).The raw data are available only on demand and consist of 14 data stripes (one data stripe for eachFPM) multiplied by the number of multispectral bands. The standard product for Landsat 8 is theortho image, available at no cost for the end-user. For PEM extractions, the advantages of using thestandard Landsat products instead of the raw data mainly include the fast -ready to use- availabilityof the data, free to registered users, which is of major importance during volcanic crises. In thisstudy, we adapt the methodology described in [21] to standard Landsat-8 products, with the aim ofsimplifying the procedure for routine use, thus widening the usability of this method for producingPEM maps. In this study, the procedure will be applied to the standard Landsat 8 data collected duringthe 26 October 2013 Mt. Etna eruptive episode and the results compared with those obtained usingdifferent satellites systems.

The paper is organized as follows: Section 1 outlines the 26 October 2013 Etna eruption andSection 2 describes the PEM procedure applied to the standard Landsat-8 products. In Section 3the results obtained are compared with the VCTH retrieved from the PEM procedure applied tothe estimations realized using the spinning enhanced visible and infrared imager (SEVIRI) and themoderate resolution imaging spectroradiometer (MODIS). In Sections 4 and 5 the discussion and theconclusions are presented.

The 26 October 2013 Mt. Etna Eruption

Mt. Etna activity in 2013 was characterized by a sequence of 16 episodes of intense eruptiveactivity at the summit of the volcano, fed by the New Southeast Crater [22,23]. The 26 October 2013episodes stand as the 14th of the year and the 39th paroxysm episode of the sequence started earlier in2011 [24–26]. The eruption occurred after a few months of quiescence and started in the early morningon 25 October, displaying mild intra-crater Strombolian activity. On 26 October, the eruptive activitygradually increased in magnitude and frequency of explosions, and lava started pouring from thecrater slowly expanding towards the Valle del Bove. In the early morning of 26 October, the explosionintensity increased markedly, and between 2:00 and 10:00 UTC the activity climaxed into a lava

Remote Sens. 2019, 11, 785 3 of 12

fountain. Over the paroxysm, the height of the lava fountain steadily reached ~200 m above the craterrim. A significant emission of gas, ash, and lapilli formed an eruptive column that rose convectivelyseveral kilometres above the summit of the volcano. The eruption ceased progressively in the lateevening, and marked ash fall was reported to be dispersed by the wind southwest of the volcanoproximally and distally down to the Ionian Mediterranean Sea [27]. Figure 1 shows the OrthorectifiedLandsat 8 image collected the 26 October 2013 at 09:37 UTC. The volcanic cloud is clearly visible.Remote Sens. 2019, 11 FOR PEER REVIEW 3

95

Figure 1. Orthorectified Landsat 8 data acquired the 26 October 2013 at 09:37 UTC on Mt. Etna volcano 96 (image courtesy of National Aeronautics and Space Administration (NASA)/United States Geological 97 Survey (USGS)). 98

2. Materials and Methods 99

The Landsat 8 OLI is a push-broom (linear array) imaging system that collects visible, near infra-100 red, and short-wave infra-red spectral band imagery at 30 m multi-spectral and 15 m panchromatic 101 ground sample distances. It collects 190 km wide image swaths from ~705 km orbital altitude [28]. 102

The OLI focal plane layout is very well described in [28] and [29]. The OLI detectors are 103 distributed across 14 separate FPMs, each of which covers a portion of the 15° OLI cross-track field 104 of view. Adjacent FPMs are offset in the along-track direction to allow for FPM-to-FPM overlap, 105 avoiding any gaps in the cross-track coverage. The internal layout of all 14 FPMs is the same, with 106 alternate FPMs being rotated by 180° to keep the active detector areas as close together as possible. 107 This feature has the effect of inverting the along-track order of the spectral bands in adjacent FPMs. 108 Consequently, this has the effect of inverting the signs of the cross-correlation measurements when 109 calculating pixels offsets between PAN and MS bands, related to the volcanic cloud velocity and 110 parallax. 111

The general concept of PEM methodology is that the PAN and the MS sensors on board a satellite 112 platform cannot occupy the same position in the focal plane of the push-broom instrument. There is 113 a physical separation between them. This separation yields a baseline and a time lag between the 114 PAN and MS image acquisitions. Two directions are considered: The epipolar direction (EP), i.e. the 115 azimuth direction of the satellite or the flight direction, and the perpendicular (P2E) to the EP 116 direction. The pixel offset between PAN and MS in the EP direction is proportional to the height of 117 the plume plus the pixel offset contribution induced by the motion of the plume itself in between the 118 two acquisitions. The pixel offsets in the P2E direction, also controlled by the time lag, are 119 proportional to the plume motion only, as there is no parallax in the P2E direction by definition. In 120

Figure 1. Orthorectified Landsat 8 data acquired the 26 October 2013 at 09:37 UTC on Mt. Etna volcano(image courtesy of National Aeronautics and Space Administration (NASA)/United States GeologicalSurvey (USGS)).

2. Materials and Methods

The Landsat 8 OLI is a push-broom (linear array) imaging system that collects visible, near infra-red,and short-wave infra-red spectral band imagery at 30 m multi-spectral and 15 m panchromatic groundsample distances. It collects 190 km wide image swaths from ~705 km orbital altitude [28].

The OLI focal plane layout is very well described in [28] and [29]. The OLI detectors are distributedacross 14 separate FPMs, each of which covers a portion of the 15◦ OLI cross-track field of view.Adjacent FPMs are offset in the along-track direction to allow for FPM-to-FPM overlap, avoiding anygaps in the cross-track coverage. The internal layout of all 14 FPMs is the same, with alternate FPMsbeing rotated by 180◦ to keep the active detector areas as close together as possible. This feature hasthe effect of inverting the along-track order of the spectral bands in adjacent FPMs. Consequently, thishas the effect of inverting the signs of the cross-correlation measurements when calculating pixelsoffsets between PAN and MS bands, related to the volcanic cloud velocity and parallax.

The general concept of PEM methodology is that the PAN and the MS sensors on board a satelliteplatform cannot occupy the same position in the focal plane of the push-broom instrument. There

Remote Sens. 2019, 11, 785 4 of 12

is a physical separation between them. This separation yields a baseline and a time lag between thePAN and MS image acquisitions. Two directions are considered: The epipolar direction (EP), i.e., theazimuth direction of the satellite or the flight direction, and the perpendicular (P2E) to the EP direction.The pixel offset between PAN and MS in the EP direction is proportional to the height of the plume plusthe pixel offset contribution induced by the motion of the plume itself in between the two acquisitions.The pixel offsets in the P2E direction, also controlled by the time lag, are proportional to the plumemotion only, as there is no parallax in the P2E direction by definition. In principle, the offset in theP2E direction is proportional to movements of every feature in the imaged scene (e.g., meteorologicalclouds, lahars, rivers flow, ocean waves, vehicles). In our case study, we are interested in the volcaniccloud motion only. We use this latter information to compensate for the apparent parallax recorded inthe EP offset.

If the data are downloaded in a staggered and orthorectified geometry, the processing is notstraightforward, since FPMs are rotated 180◦, and the offset analysis by cross-correlation would yieldopposite signs at adjacent image stripes. This hampers the correct deployment of the method describedin [21]. To avoid this inconvenience, we propose the following 5 step procedure:

I. The dataset is rotated, so that the columns of the image matrix are aligned to the nominalazimuth direction of the satellite reported in the ancillary data files.

II. A correlator to perform pixel (or sub-pixel) offset measurements is used (e.g., [30]). If oneconsiders the OLI image as a matrix made of lines and columns, offsets among lines are the EPoffsets (Oe), while offsets among columns are the P2E offsets (Op2e). In the offset results, therecould exist a ramp resulting from band mis-registration; the ramp, if found, is removed.

III. The direction of the plume is measured with respect to the azimuth direction, with theconvention depicted in Figure 2.

IV. The absolute value of the pixel offsets due to the VCTH is calculated as Oh. Generally, Oh fromstaggered sensors should be calculated as follows:

|Oh| = |Oe| −∣∣Op2e

∣∣ |tan θ| (1)

if theta is between zero and 180, or

|Oh| = |Oe|+∣∣Op2e

∣∣ |tan θ| (2)

if theta is between 180 and 360. Oh is then converted into a VCTH using the formula providedin [20]:

h = |Oh|.s.HV.t

(3)

for every pixel, which makes it a PEM. h is the plume height (m), s is the pixel size (m), V isthe platform velocity (m/s), t is the temporal lag between the two Landsat 8 bands (s) and His the platform height (m). Peculiar cases are: θ = 0◦ and θ = 180◦. In these cases, the system isno longer sensitive to plume velocity. Therefore,

|Oh| = |Oe| (4)

V. Finally, the results are re-rotated to their original position. Then, one has to choose a knownreference altitude value on land and attribute it to the corresponding pixel. In our case study,we choose to set to zero the coastline close to the city of Catania.

Remote Sens. 2019, 11, 785 5 of 12Remote Sens. 2019, 11 FOR PEER REVIEW 5

150

Figure 2. Angles and geometrical conventions, imagining this is a rotated Landsat 8 scene with a 151 volcano at its centre. The epipolar (flight motion) direction and the perpendicular (p2e) to the 152 epipolar direction (EP) direction are indicated. Dark gray and light gray colours indicate 153 quadrants where the offsets are either summed up, either subtracted respectively. Ovals represent 154 possible ash clouds directions. The light blue colour indicate the ash cloud direction of the case 155 study presented here. 156

Figure 3 shows the offsets results. We show the raw results of the correlator on the left sides and 157 the corrected results on the right side. The vertical stripes on the left sides are due to volcanic (and 158 non-volcanic) cloud velocities and parallax: As the FPMs are inverted 180°, the correlator yields 159 velocities with sign opposition (as explained in the introduction). The correlation results are corrected 160 by using |𝑂𝑒| , |𝑂𝑝2𝑒| as described in the above paragraphs. The pixel offsets are expressed in meters. 161

It is interesting to note that the P2E offset correspond to cloud (volcanic cloud and non-volcanic 162 cloud) velocities, which values are comparable to the wind speed represented in Figure 6. 163

Figure 2. Angles and geometrical conventions, imagining this is a rotated Landsat 8 scene with avolcano at its centre. The epipolar (flight motion) direction and the perpendicular (p2e) to the epipolardirection (EP) direction are indicated. Dark gray and light gray colours indicate quadrants wherethe offsets are either summed up, either subtracted respectively. Ovals represent possible ash cloudsdirections. The light blue colour indicate the ash cloud direction of the case study presented here.

Figure 3 shows the offsets results. We show the raw results of the correlator on the left sidesand the corrected results on the right side. The vertical stripes on the left sides are due to volcanic(and non-volcanic) cloud velocities and parallax: As the FPMs are inverted 180◦, the correlator yieldsvelocities with sign opposition (as explained in the introduction). The correlation results are correctedby using |Oe|,

∣∣Op2e∣∣ as described in the above paragraphs. The pixel offsets are expressed in meters.

It is interesting to note that the P2E offset correspond to cloud (volcanic cloud and non-volcanic cloud)velocities, which values are comparable to the wind speed represented in Figure 6.

Remote Sens. 2019, 11, 785 6 of 12

Remote Sens. 2019, 11 FOR PEER REVIEW 6

164

Figure 3. (a): Results from the correlator show vertical stripes due to volcanic (and non-volcanic) cloud 165 velocities. As focal plane modules (FPMs) are inverted 180°, the correlator yields velocities with sign 166 opposition (see text for more details). (b): Correlation results are corrected by using |𝑂𝑒| , |𝑂𝑝2𝑒| as 167 described in this study. EP offsets (Oe), P2E offsets (Op2e). The pixel offset is expressed in meters. It 168 is interesting to note that the P2E offset correspond to cloud velocities. 169

3. Results and cross-comparisons 170

Figure 4 shows the VCTH map obtained from the PEM procedure applied to the Landsat 171 orthorectified data. The volcanic cloud height vary from about 6 up to 9.5 km above the sea level 172 (a.s.l.) with the higher values that lie in the central region of the cloud. 173

The estimated VCTH has been compared the VCTH extracted by using different procedure 174 applied to other satellite sensors. Note that the cross-comparison here is not used as a validation. It 175 allows us to assess the consistency of the results (i. e. our results are in the same order/scale as 176 independent measurements). A thorough validation is not possible since acquisition times (repeat 177 cycle) of different sensors are not the same as Landsat. Since the VCTH evolves with time, we prefer 178 to call it « cross-comparisons » rather than « validation ». In addition, different sensors acquire data 179 from different positions, introducing a bias in the eventual validation campaign. 180

Figure 3. (a): Results from the correlator show vertical stripes due to volcanic (and non-volcanic) cloudvelocities. As focal plane modules (FPMs) are inverted 180◦, the correlator yields velocities with signopposition (see text for more details). (b): Correlation results are corrected by using |Oe|,

∣∣Op2e∣∣ as

described in this study. EP offsets (Oe), P2E offsets (Op2e). The pixel offset is expressed in meters. It isinteresting to note that the P2E offset correspond to cloud velocities.

3. Results and Cross-Comparisons

Figure 4 shows the VCTH map obtained from the PEM procedure applied to the Landsatorthorectified data. The volcanic cloud height vary from about 6 up to 9.5 km above the sea level (a.s.l.)with the higher values that lie in the central region of the cloud.

The estimated VCTH has been compared the VCTH extracted by using different procedureapplied to other satellite sensors. Note that the cross-comparison here is not used as a validation.It allows us to assess the consistency of the results (i.e., our results are in the same order/scale asindependent measurements). A thorough validation is not possible since acquisition times (repeatcycle) of different sensors are not the same as Landsat. Since the VCTH evolves with time, we prefer tocall it « cross-comparisons » rather than « validation ». In addition, different sensors acquire data fromdifferent positions, introducing a bias in the eventual validation campaign.

Here the VCTH of the Etna 26 October 2013 eruption, used for the cross-comparison with Landsatresults, are estimated by using geostationary (SEVIRI) and polar (MODIS) satellite sensors.

Remote Sens. 2019, 11, 785 7 of 12Remote Sens. 2019, 11 FOR PEER REVIEW 7

181

Figure 4. The volcanic cloud elevation (km), extracted from Landsat 8 orthorectified image. 182

Here the VCTH of the Etna 26 October 2013 eruption, used for the cross-comparison with 183 Landsat results, are estimated by using geostationary (SEVIRI) and polar (MODIS) satellite sensors. 184

3.1. MODIS VCTH estimation 185

MODIS is a multispectral radiometer on board the NASA Terra and Aqua polar satellites. It has 186 36 spectral channels from visible (VIS) to thermal infrared (TIR), with a spatial resolution at sub-187 satellite of 1 km in the TIR and repetition cycle of 1–2 days. The VCTH is computed by exploiting the 188 well known “dark pixels” procedure, based on the comparison between the brightness temperature 189 at 11 m of the coldest volcanic cloud pixel (Tb,11), with the atmospheric temperature profile of the 190 same region at same time [2]. The temperature profile has been obtained from the National Centers 191 for Environmental Prediction (NCEP)/National Center for Atmospheric Research (NCAR) [31], 192 considering a box with 2.5° x 2.5° centred on Etna at 12:00 UTC. The left panel of Figure 5 shows the 193 Tb,11 for the MODIS-Terra image collected the 26 October 2013 at 09:00 UTC. The volcanic cloud is 194 clearly visible as dark signature on the right side of the image, while the cyan region indicates no 195 data. The right panel of Figure 5 shows the NCEP temperature profile (grey line) and the Tb,11 of the 196 dark pixel (red vertical line). Being Tb,11 = -37.4 °C, VCTH result 8.9 km with an uncertainty of +/- 500 197 m, computed considering Tb,11 +/- 2 °C [2,32]. 198

Figure 4. The volcanic cloud elevation (km), extracted from Landsat 8 orthorectified image.

3.1. MODIS VCTH Estimation

MODIS is a multispectral radiometer on board the NASA Terra and Aqua polar satellites. It has 36spectral channels from visible (VIS) to thermal infrared (TIR), with a spatial resolution at sub-satellite of1 km in the TIR and repetition cycle of 1–2 days. The VCTH is computed by exploiting the well known“dark pixels” procedure, based on the comparison between the brightness temperature at 11 µm of thecoldest volcanic cloud pixel (Tb,11), with the atmospheric temperature profile of the same region atsame time [2]. The temperature profile has been obtained from the National Centers for EnvironmentalPrediction (NCEP)/National Center for Atmospheric Research (NCAR) [31], considering a box with2.5◦ × 2.5◦ centred on Etna at 12:00 UTC. The left panel of Figure 5 shows the Tb,11 for the MODIS-Terraimage collected the 26 October 2013 at 09:00 UTC. The volcanic cloud is clearly visible as dark signatureon the right side of the image, while the cyan region indicates no data. The right panel of Figure 5shows the NCEP temperature profile (grey line) and the Tb,11 of the dark pixel (red vertical line). BeingTb,11 = −37.4 ◦C, VCTH result 8.9 km with an uncertainty of +/− 500 m, computed considering Tb,11+/− 2 ◦C [2,32].

Remote Sens. 2019, 11, 785 8 of 12

Remote Sens. 2019, 11 FOR PEER REVIEW 8

Figure 5. Left panel: brightness temperature at 11 m for the moderate resolution imaging 199 spectroradiometer (MODIS)-Terra image collected the 26 October 2013 at 09:00 UTC. Right panel: 200 National Centers for Environmental Prediction (NCEP) temperature profile (grey line) and dark pixel 201 brightness temperature (red vertical line). 202

3.2. SEVIRI VCTH estimation 203

The SEVIRI instrument on board METEOSAT Second Generation (MSG) geostationary satellites 204 is a 12 channel VIS-TIR multi-channel imager, which operates from 15 minutes repeat cycle on entire 205 hemisphere (Full Disk) to 5 minutes over Europe (Rapid Scan). The sub-satellite point spatial 206 resolution is 3 × 3 km2, and the pixel dimension in the Etnean area is about 4.3 × 3.3 km. Exploiting 207 the high data frequency of the SEVIRI images, the volcanic cloud speed can be retrieved by following 208 the volcanic cloud centre of mass. By making a basic assumption that the estimated centre of mass 209 speed is the whole volcanic cloud speed, VCTH can be obtained by comparing this value with the 210 wind speed profile collected in the same time and position [26]. Also in this case the wind profile 211 derive from NCEP/NCAR considering a box with 2.5° x 2.5° centred on Etna at 12:00 UTC. The upper 212 panels of Figure 6 show the volcanic cloud ash mass maps obtained from the SEVIRI images collected 213 at 09:00 UTC (left panel) and at 10:00 UTC (right panel). In these two images the different position of 214 the volcanic cloud centre of mass is clearly identifiable. The distance of the centre of mass from the 215 vents and the time of acquisition of the SEVIRI images, allows the computation of the volcanic cloud 216 speed (see lower-left panel). This retrieved value is then compared with the wind speed NCEP/NCAR 217 profile (see lower-right panel). In this case the wind speed of the volcanic cloud centre of mass is 18.0 218 m/s that yield to a VCTH of 10.5 km. An uncertainty of +/- 500 m is associated to take into account 219 the uncertainty in the centre of mass identification. It is interesting to note that the P2E offset in Figure 220 3 corresponding to volcanic cloud velocities, is comparable to the wind speed measured by 221 NCEP/NCAR (~18 m/s) shown in Figure 6. 222

223

Figure 5. Left panel: brightness temperature at 11 µm for the moderate resolution imagingspectroradiometer (MODIS)-Terra image collected the 26 October 2013 at 09:00 UTC. Right panel:National Centers for Environmental Prediction (NCEP) temperature profile (grey line) and dark pixelbrightness temperature (red vertical line).

3.2. SEVIRI VCTH Estimation

The SEVIRI instrument on board METEOSAT Second Generation (MSG) geostationary satellitesis a 12 channel VIS-TIR multi-channel imager, which operates from 15 min repeat cycle on entirehemisphere (Full Disk) to 5 min over Europe (Rapid Scan). The sub-satellite point spatial resolution is3 × 3 km2, and the pixel dimension in the Etnean area is about 4.3 × 3.3 km. Exploiting the high datafrequency of the SEVIRI images, the volcanic cloud speed can be retrieved by following the volcaniccloud centre of mass. By making a basic assumption that the estimated centre of mass speed is thewhole volcanic cloud speed, VCTH can be obtained by comparing this value with the wind speedprofile collected in the same time and position [26]. Also in this case the wind profile derive fromNCEP/NCAR considering a box with 2.5◦ x 2.5◦ centred on Etna at 12:00 UTC. The upper panelsof Figure 6 show the volcanic cloud ash mass maps obtained from the SEVIRI images collected at09:00 UTC (left panel) and at 10:00 UTC (right panel). In these two images the different position ofthe volcanic cloud centre of mass is clearly identifiable. The distance of the centre of mass from thevents and the time of acquisition of the SEVIRI images, allows the computation of the volcanic cloudspeed (see lower-left panel). This retrieved value is then compared with the wind speed NCEP/NCARprofile (see lower-right panel). In this case the wind speed of the volcanic cloud centre of mass is18.0 m/s that yield to a VCTH of 10.5 km. An uncertainty of +/- 500 m is associated to take intoaccount the uncertainty in the centre of mass identification. It is interesting to note that the P2E offsetin Figure 3 corresponding to volcanic cloud velocities, is comparable to the wind speed measured byNCEP/NCAR (~18 m/s) shown in Figure 6.

Remote Sens. 2019, 11, 785 9 of 12

Remote Sens. 2019, 11 FOR PEER REVIEW 9

Figure 6. Upper panels: volcanic cloud ash mass maps obtained from the spinning enhanced visible 224 and infrared imager (SEVIRI) images collected at 09:00 UTC (left panel) and at 10:00 UTC (right 225 panel). Lower panels: Computation of the volcanic cloud speed from SEVIRI images (left panel) and 226 comparison between the value obtained and the wind speed NCEP/ National Center for Atmospheric 227 Research (NCAR) profile (right panel). 228

4. Discussion 229

The simple cross comparison presented here indicates that VCTH retrievals from Landsat 8, 230 SEVIRI, and MODIS are in good agreement, taking into account the different methodologies and the 231 different times of images acquisition (9:37 UTC for Landsat, 9:00 UTC for MODIS and from 9 to 10:30 232 UTC for SEVIRI). The volcanic plume is a complex medium made of different materials and chemical 233 species. Different materials (such as ash or ice) and different chemicals (such as SO2, H2O, and CO2, 234 and halogens) are dispersed at different altitudes. As an example, sensors that capture or model SO2 235 dispersion in the atmosphere (such as the Infrared Atmospheric Sounding Interferometer, Iasi) will, 236 therefore, measure a different plume altitude compared to sensors that measure visible plume 237 particles (such as Landsat 8 and SEVIRI). This important distinction has to be taken into account 238 when comparing results from different sensors. The PEM method works best for volcanic clouds 239 made of ash particles and optically thick material. 240

The accuracy of the PEM extracted from Landsat 8 has been already assessed against ground 241 cameras in de Michele et al. (2016) for the Holuraun fissural eruption (Iceland). They reported an 242 accuracy of 300 m. Landsat 8 acquires data at a high spatial resolution (15/30 m grid) at the cost of a 243 medium revisit time (every 14 days). Instead, SEVIRI has the advantage of acquiring data every half 244 an hour everywhere in the globe. This is done at the cost of medium spatial resolution (1 km grid). 245 Statistics performed over VCTH retrieved by Landsat 8 (PEM), SEVIRI and MODIS data on a 246 common area show very good agreement with a maximum height value of 8 (with 1 km precision). 247 The strength in the results show the complementarity between these systems. Assimilation models 248 could benefit from complementary constraints brought by these sensors. 249

Some questions remain open to future studies. How large must the optical depth or 250 concentration of the particles be to enable VCTH-assessment? How far from the source is the retrieval 251 of VCTH typically possible with Landsat 8 ? We believe that this information would be essential for 252

Figure 6. Upper panels: volcanic cloud ash mass maps obtained from the spinning enhanced visibleand infrared imager (SEVIRI) images collected at 09:00 UTC (left panel) and at 10:00 UTC (right panel).Lower panels: Computation of the volcanic cloud speed from SEVIRI images (left panel) andcomparison between the value obtained and the wind speed NCEP/ National Center for AtmosphericResearch (NCAR) profile (right panel).

4. Discussion

The simple cross comparison presented here indicates that VCTH retrievals from Landsat 8,SEVIRI, and MODIS are in good agreement, taking into account the different methodologies and thedifferent times of images acquisition (9:37 UTC for Landsat, 9:00 UTC for MODIS and from 9 to 10:30UTC for SEVIRI). The volcanic plume is a complex medium made of different materials and chemicalspecies. Different materials (such as ash or ice) and different chemicals (such as SO2, H2O, and CO2,and halogens) are dispersed at different altitudes. As an example, sensors that capture or modelSO2 dispersion in the atmosphere (such as the Infrared Atmospheric Sounding Interferometer, Iasi)will, therefore, measure a different plume altitude compared to sensors that measure visible plumeparticles (such as Landsat 8 and SEVIRI). This important distinction has to be taken into account whencomparing results from different sensors. The PEM method works best for volcanic clouds made ofash particles and optically thick material.

The accuracy of the PEM extracted from Landsat 8 has been already assessed against groundcameras in de Michele et al. (2016) for the Holuraun fissural eruption (Iceland). They reported anaccuracy of 300 m. Landsat 8 acquires data at a high spatial resolution (15/30 m grid) at the cost of amedium revisit time (every 14 days). Instead, SEVIRI has the advantage of acquiring data every halfan hour everywhere in the globe. This is done at the cost of medium spatial resolution (1 km grid).Statistics performed over VCTH retrieved by Landsat 8 (PEM), SEVIRI and MODIS data on a commonarea show very good agreement with a maximum height value of 8 (with 1 km precision). The strengthin the results show the complementarity between these systems. Assimilation models could benefitfrom complementary constraints brought by these sensors.

Some questions remain open to future studies. How large must the optical depth or concentrationof the particles be to enable VCTH-assessment? How far from the source is the retrieval of VCTH

Remote Sens. 2019, 11, 785 10 of 12

typically possible with Landsat 8 ? We believe that this information would be essential for designing anew, dedicated mission. In general, in our approach, the plume height is not measureable if the plumeis not visible in the image (Table 1).

Table 1. Summary of the comparison between the different volcanic cloud-top height (VCTH) retrievalsderived from Landsat 8, MODIS and SEVIRI.

VCTH [km] VCTH Uncertainty

LANDSAT 9.93 0.3MODIS 8.9 0.5SEVIRI 10.5 0.5

5. Conclusions

Landsat 8 satellites offer a unique opportunity for monitoring volcanic plumes at a high spatialresolution together with the opportunity to study volcanic eruptions back in time by exploiting the dataarchives. De Michele et al. [21] presented a method to extract the PEM (i.e., the digital elevation modelof a volcanic plume) from raw Landsat 8 data. Nevertheless, raw Landsat 8 data are not availableto the general public, limiting de facto the straight-forward application of the method. Moreover,the Landsat 8 OLI sensor presents a staggered geometry, which needs tuned processing. In this study,we push the methodology forward: Here we present a generalized, simplified methodology to extractthe VCTH as a PEM from the common push broom staggered sensors, such as the standard Landsat 8products. The advantages are manifold. First, it takes advantage of the freely available Landsat 8 dataarchives. Second, the method exploits the already-orthorectified Landsat 8 dataset, which is a standardLandsat 8 product, readily available on Landsat 8 archives. Third, the method could be adapted to theCopernicus Sentinel 2 data, a staggered sensor for which the products available to general public areorthorectified (similar to Landsat 8). The synergetic use of multiple sensors could improve the revisittime over a given volcanic eruption. It represents a step towards the routine monitoring of volcanicplumes height from space, at high spatial resolution.

The VCTH retrievals obtained from the geostationary SEVIRI and the polar MODIS satelliteinstruments indicate a good agreement with the Landsat VCTH product.

Funding: This study benefited from funding of the European Union and BRGM within the APHORISM (AdvancedProcedures for Volcanic and Seismic Monitoring) project, 7th framework programme of the European Union.We are thankful to NASA/USGS Landsat 8 program for the Landsat 8 OLI data.

Acknowledgments: We are thankful to NASA/USGS for the Landsat8 data.

Conflicts of Interest: The authors declare no conflict of interest.

References

1. Wen, S.; Rose, W.I. Retrieval of sizes and total masses of particles in volcanic clouds using AVHRR bands 4and 5. J. Geophys. Res. Atmos. 1994, 99, 5421–5431. [CrossRef]

2. Prata, A.J.; Grant, I.F. Retrieval of microphysical and morphological properties of volcanic ash plumesfrom satellite data: Application to Mt Ruapehu, New Zealand. Q. J. R. Meteorol. Soc. 2001, 127, 2153–2179.[CrossRef]

3. Corradini, S.; Merucci, L.; Prata, A.J. Retrieval of SO2 from thermal infrared satellite measurements:Correction procedures for the effects of volcanic ash. Atmos. Meas. Tech. 2009, 2, 177–191. [CrossRef]

4. Mastin, L.G.; Guffanti, M.; Servranckx, R.; Webley, P.; Barsotti, S.; Dean, K.; Durant, A.; Ewert, J.W.; Neri, A.;Rose, W.I.; et al. A multidisciplinary effort to assign realistic source parameters to models of volcanicash-cloud transport and dispersion during eruptions. J. Volcanol. Geotherm. Res. 2009, 186, 10–21. [CrossRef]

5. Stohl, A.; Prata, A.J.; Eckhardt, S.; Clarisse, L.; Durant, A.; Henne, S.; Kristiansen, N.I.; Minikin, A.;Schumann, U.; Seibert, P.; et al. Determination of time- and height-resolved volcanic ash emissions and theiruse for quantitative ash dispersion modeling: The 2010 Eyjafjallajökull eruption. Atmos. Chem. Phys. 2011,11, 4333–4351.

Remote Sens. 2019, 11, 785 11 of 12

6. Poret, M.; Corradini, S.; Merucci, L.; Costa, A.; Andronico, D.; Montopoli, M.; Vulpiani, G.; Freret-Lorgeril, V.Reconstructing volcanic plume evolution integrating satellite and ground-based data: Application to the23rd November 2013 Etna eruption. Atmos. Chem. Phys. Discuss. 2018. [CrossRef]

7. Richards, M.S. Volcanic Ash Cloud Heights Using the MODIS CO2-Slicing Algorithm. Master’s Thesis,University of Wisconsin, Madison, WI, USA, 6 January 2006.

8. Chang, F.-L.; Minnis, P.; Lin, B.; Khaiyer, M.M.; Palikonda, R.; Spangenberg, D.A. A modified methodfor inferring upper troposphere cloud top height using the GOES 12 imager 10.7 and 13.3 µm data.J. Geophys. Res. 2010, 115, D06208. [CrossRef]

9. Corradini, S.; Cervino, M. Aerosol extinction coefficient profile retrieval in the Oxygen A-band consideringmultiple scattering atmosphere. Test case: SCIAMACHY nadir simulated measurements. J. Quant. Spectrosc.Radiat. Transf. 2006, 97, 354–380. [CrossRef]

10. Dubuisson, P.; Frouin, R.; Dessailly, D.; Duforêt, L.; Léon, J.-F.; Voss, K.; Antoine, D. Estimating the altitude ofaerosol plumes over the ocean from reflectance ratio measurements in the O2 A-band. Remote Sens. Environ.2009, 113, 1899–1911. [CrossRef]

11. Biondi, R.; Steiner, A.K.; Kirchengast, G.; Brenot, H.; Rieckh, T. Supporting the detection and monitoring ofvolcanic clouds: A promising new application of Global Navigation Satellite System radio occultation. Adv.Space Res. 2017, 60, 2707–2722. [CrossRef]

12. Pardini, F.; Burton, M.; de’ Michieli Vitturi, M.; Corradini, S.; Salerno, G.; Merucci, L.; Di Grazia, G.Retrieval and intercomparison of volcanic SO2 injection height and eruption time from satellite maps andground-based observations. J. Volcanol. Geotherm. Res. 2017. [CrossRef]

13. Merucci, L.; Zakšek, K.; Carboni, E.; Corradini, S. Steeoscopic estimation of volcanic cloud-top height fromtwo geostationary satellites. Remote Sens. 2016, 8, 206. [CrossRef]

14. Winker, D.M.; Liu, Z.; Omar, A.; Tackett, J.; Fairlie, D. CALIOP observations of the transport of ash from theEyjafjallajökull volcano in April 2010. J. Geophys. Res. 2012, 117, D00U15. [CrossRef]

15. Prata, A.J.; Turner, P.J. Cloud top height determination from the ATSR. Remote Sens. Environ. 1997, 59, 1.[CrossRef]

16. Mims, S.R.; Kahn, R.A.; Moroney, C.M.; Gaitley, B.J.; Nelson, D.L.; Garay, M.J. MISR stereo heights ofgrassland fire smoke plumes in Australia. IEEE Trans. Geosci. Remote Sens. 2010, 48, 25–35. [CrossRef]

17. Nelson, D.L.; Garay, M.J.; Kahn, R.A.; Dunst, B.A. Stereoscopic Height and Wind Retrievals for AerosolPlumes with the MISR INteractive eXplorer (MINX). Remote Sens. 2013, 5, 4593–4628. [CrossRef]

18. Flower, V.J.; Kahn, R.A. Assessing the altitude and dispersion of volcanic plumes using MISR multi-angleimaging from space: Sixteen years of volcanic activity in the Kamchatka Peninsula, Russia. J. Volcanol.Geotherm. Res. 2017, 337, 1–15. [CrossRef]

19. Zakšek, K.; Hort, M.; Zaletelj, J.; Langmann, B. Monitoring volcanic ash cloud top height throughsimultaneous retrieval of optical data from polar orbiting and geostationary satellites. Atmos. Chem. Phys.2013, 13, 2589–2606. [CrossRef]

20. Corradini, S.; Montopoli, M.; Guerrieri, L.; Ricci, M.; Scollo, S.; Merucci, L.; Marzano, F.S.; Pugnaghi, S.;Prestifilippo, M.; Ventress, L.; et al. A multi-sensor approach for the volcanic ash cloud retrievals anderuption characterization. Remote Sens. 2016, 8, 58. [CrossRef]

21. De Michele, M.; Raucoules, D.; Arason, Þ. Volcanic plume elevation model and its velocity derived fromLandsat 8. Remote Sens. Environ. 2016, 176, 219–224. [CrossRef]

22. Ferlito, C.; Bruno, V.; Salerno, G.; Caltabiano, T.; Scandura, D.; Mattia, M.; Coltorti, M. Dome-like behavior atMt. Etna: The case of the 28 December 2014 South East Crater paroxysm. Sci. Rep. 2017, 7, 5361. [CrossRef]

23. Patanè, D.; Aiuppa, A.; Aloisi, M.; Behncke, B.; Cannata, A.; Coltelli, M.; Di Grazia, G.; Gambino, S.;Gurrieri, S.; Mattia, M.; et al. Insights into magma and fluid transfer at Mount Etna by a multi parametricapproach: A model of the events leading to the 2011 eruptive cycle. J. Geophys. Res. 2013, 118, 1–21.[CrossRef]

24. Behncke, B.; Branca, S.; Corsaro, R.A.; De Beni, E.; Miraglia, L.; Proietti, C. The 2011–2012 summit activity ofMount Etna: Birth, growth and products of the new SE crater. J. Volcanol. Geotherm. Res. 2014, 270, 10–21.[CrossRef]

25. Spampinato, L.; Sciotto, M.; Cannata, A.; Cannavò, F.; La Spina, A.; Palano, M.; Salerno, G.G.; Privitera, E.;Caltabiano, T. Multiparametric study of the February–April 2013 paroxysmal phase of Mt. Etna NewSouth-East crater. Geochem. Geophys. Geosyst. 2015, 16, 1932–1949. [CrossRef]

Remote Sens. 2019, 11, 785 12 of 12

26. Corradini, S.; Guerrieri, L.; Lombardo, V.; Merucci, L.; Musacchio, M.; Prestifilippo, M.; Scollo, S.; Silvestri, M.;Spata, G.; Stelitano, D. Proximal monitoring of the 2011-2015 Etna lava fountains using MSG-SEVIRI data.Geosciences 2018, 8, 140. [CrossRef]

27. Sellitto, P.; di Sarra, A.; Corradini, S.; Boichu, M.; Herbin, H.; Dubuisson, P.; Sèze, G.; Meloni, D.;Monteleone, F.; Merucci, L.; et al. Synergistic use of Lagrangian dispersion and radiative transfer modellingwith satellite and surface remote sensing measurements for the investigation of volcanic plumes: The MountEtna eruption of 25–27 October 2013. Atmos. Chem. Phys. 2016, 16, 6841–6861. [CrossRef]

28. Storey, J.; Choate, M.; Lee, K. Landsat-8 operational land imager on-orbit geometric calibration andperformance. Remote Sens. 2014, 6, 11127–11152. [CrossRef]

29. Knight, E.; Kvaran, G. Landsat-8 operational land imager design, characterization, and performance. RemoteSens. 2014, 6, 10286–10305. [CrossRef]

30. Leprince, S.; Barbot, S.; Ayoub, F.; Avouac, J.P. Automatic and precise orthorectification, coregistration, andsubpixel correlation of satellite images, application to ground deformation measurements. IEEE Trans. Geosci.Remote Sens. 2007, 45, 6. [CrossRef]

31. The NCEP/NCAR Reanalysis Project at the NOAA/ESRL Physical Sciences Division. Available online:http://www.esrl.noaa.gov/psd/data/reanalysis/reanalysis.shtml (accessed on 1 December 2017).

32. Corradini, S.; Merucci, L.; Prata, A.J.; Piscini, A. Volcanic ash and SO2 in the 2008 Kasatochi eruption:Retrievals comparison from different IR satellite sensors. JGR 2010, 115, D00L21. [CrossRef]

© 2019 by the authors. Licensee MDPI, Basel, Switzerland. This article is an open accessarticle distributed under the terms and conditions of the Creative Commons Attribution(CC BY) license (http://creativecommons.org/licenses/by/4.0/).